ANALYSIS OF WELL WATER LEVEL TIME SERIES DATA FOR AQUIFER PARAMETER ESTIMATION

There is great interest in the subsurface inland propagation of sea level rise, tidal fluctuations, and storm surges into the Biscayne Aquifer. In addition to direct flooding from high water tables, flooding from rain will be intensified when the water table rises because the storage capacity of the soil is diminished. Also, high water tables contribute to infiltration and inflow into sanitary sewer systems, requiring the treatment of additional water that may upset treatment processes due to its chemistry.

This research examines the effects of ocean tides on the groundwater level fluctuations in the coastal aquifer. The tide effect is usually much smaller than the effect of precipitation on the water level. This study includes USGS data from 14 wells in Miami-Dade and Broward counties, NOAA data from the closest five tidal gages, and data from five tidal structures from the South Florida Water Management District’s DBHYDRO database. The well closest to the shore is 1500 feet away and the farthest well is 15000 feet away.

An analytical model for tidal impacts on groundwater has been available since at least 1950 (Jacob, 1950). For a sinusoidal tide, the model predicts that the change in the water table lags increasingly behind the tide in time as the distance inland increases, and that the amplitude of the tidal signal drops off with increasing distance inland.

If the T/S hydraulic diffusivity parameter is known, prediction of the inland effects is straightforward. Determination of the parameter is notoriously difficult in Southeast Florida however, and another means of measuring it would be very useful.

One potentially powerful strategy for using the Jacob model to estimate T/S would be to look at water level time series obtained from wells and tide gages and attempt to fit the observations to the model. Examination of cross-correlation functions is also useful to determinate the time delay between the signal of tides and inland wells. Because the strong impact of rainfall events and managed canal water levels on water table elevations complicates analysis of the well data, only data from periods with no rainfall events can be used in the simplest analysis. More advanced analysis might permit detection of the tidal signal in longer time series that also have rain.